Image-based vehicle maneuvering assistant methods and systems had been available and used for many years. However, most such conventional image-based vehicle maneuvering assistant methods and systems are only designed to display images surrounding their corresponding vehicles to drivers directly and exactly as they are being captured in their image sensing devices. Thus, in despite of they actually are helpful with respect to driving assistance, they can sometimes create a kind of visual burden to drivers with the increasing of image sensing devices since there can be a plurality of images from the image sensing devices crowding in one monitoring screen. For overcoming such problem, it is in need of a technique for constructing a surrounding seamless bird-view image by the use of a plurality of surrounding images captured respectively from different image sensing devices. Such technique for constructing a full image in the bird's eye view surrounding a vehicle is becoming the mainstream in the automobile manufacturing industry when it come to the developing of a more advance and safer vehicle. Operationally, the surrounding images from common image sensing devices installed on various locations of a vehicle are processed and converted into corresponding bird-view images which are then being combined into a full image in the bird's eye view surrounding the vehicle to be provided to the vehicle's driver, and thereby, the driving safety of the vehicle enhanced.
However, the key issue for combining the plural surrounding images captured respectively by different image sensing devices located at different positions on a vehicle is to recognize and determine if there is an overlapped region in two surrounding images captured by any two image sensing devices that are adjacent to each other. One method for dealing with the aforesaid key issue is disclosed in U.S. Pat. Pub. No. 2007/0003108, which reveals a technique for combing a plurality of images taken by plural cameras focusing in different directions around a vehicle into a bird-view image showing the entire view surrounding the vehicle. However, as the bird-view image is divided into plural regions by setting mask line at each joining line between adjacent images, the bird-view image in this disclosure is not seamless combined. That is, the various surrounding images taken by the plural cameras in different directions around the vehicle are not combined seamlessly that there can be overlapped region existed in two images taken by any two adjacent cameras, and therefore, it may required to have ultrasonic radar detectors to be arranged on the vehicle at positions corresponding to the overlapped regions as alarming devices for assisting the maneuvering of the vehicle. Nevertheless, the arranging of the ultrasonic radar detectors is going to cost the manufacturing cost of the aforesaid system to increase.
Moreover, in a driving support system disclosed in U.S. Pat. Pub. No. 2006/0202984, different all-round bird-view images are generated from the images taken by four cameras mounted on a vehicle according to the case where the vehicle is stopped, the case where the vehicle is driven forward, and the case where the vehicle is driven in reverse while the depression angle of each camera for capturing images is defined by the height between the corresponding camera and the ground, by that the system is able to simulate and acquire a larger all-round bird-view image showing the entire view surrounding the vehicle. In Japanese Patent Laid-Open Publication No. 2008-048317, a method for weighting corresponding features in the overlapped region between two images taken by two adjacent cameras that are mounted on a carrier, by which the two images can be combined in a manner that the features with higher weights are preserved in the overlapped region while the corresponding feature with smaller weight in the overlapped region are deleted, and thereby, various surrounding images taken by the plural cameras in different directions around the carrier can be combined into a bird-view image showing the entire view surrounding the carrier. In U.S. Pat. Pub. No. 2009/0097708, an image-processing system for processing image data obtained by capturing a plurality of images outside a periphery of a vehicle is provided, in which a plurality of image-capturing units, being affixed to the vehicle, are used for generating bird-view images outside the periphery of the vehicle in a manner that end portions of real spaces corresponding to two adjacent bird-view images overlap each other, and accordingly the generated bird-view images are provided to an image-composition unit where they are combined into a bird-view-composite image according to a predetermined layout as each overlap portion being included are determined to be a border image that is provided to be superimposed on its corresponding bird-view image. Thus, by the composition of those bird-view images with borders, a surrounding bird-view image showing the entire view surrounding the vehicle can be achieved. In Japanese Patent Laid-Open Publication No. 2006-121587, an operation supporting device capable of suppressing deterioration in image quality due to the joint between images is provided, by which the device changes the joint position of images in image synthesis when the bird-view image is created to a position different from the previous one. Therefore, the joint does not continuously exist at a fixed position in the bird-view image, thus preventing an object from being displayed discontinuously at the joint between the images, preventing the object from continuing to disappear, and hence suppressing the deterioration of the image quality caused by the joint between the images. It is noted that the key issues in the abovementioned four patents for combining the plural surrounding images captured respectively by different image sensing devices located at different positions on a vehicle are to recognize and determine if there is an overlapped region in two surrounding images captured by any two image sensing devices that are adjacent to each other. Then, as each surrounding image is transformed into a corresponding bird-view image with reference to the location of its corresponding image sensing devices, the plural bird-view images can be combined with reference to the identified overlapped regions so as to layout and achieve a surrounding bird-view image showing the entire view surrounding the vehicle.
In addition, in U.S. Pat. No. 7,161,616, an image processing apparatus is provided, which includes an image processing part for receiving images captured by a plurality of cameras shooting surroundings of a vehicle to generate a synthetic image from these camera images, wherein the image processing part uses mapping data describing a correspondence relationship between pixels of the synthetic image and a plurality of pixel data including one or both of pixel data of the camera images and pixel data other than the camera images, and describing the rate of necessity with respect to each of the pixel data, and weights each pixel data in accordance with the rate of necessity, thereby generating the pixel data of the pixels of the synthetic image.